One Immune Simplex Particle Swarm Optimization and It's Application

In order to improve the performance of particle swarm optimization, one immune simplex PSO is proposed in the paper. Firstly, orthogonal principle is applied to the initialization of particle, so particles are evenly distributed. Secondly, during the evolution, clone and mutation operation based on fitness evaluation is applied to the particle under the concept of immune. Thirdly, because the value of velocity is close to 0 when the particle is in local optimization, the velocity of the particle is also modified after resetting the position according to the simplex principle. It is verified by multidimensional, nonlinear function (Grievank function) in MATLAB. The result shows that the improved PSO can realize global optimization effectively. And in order to analyze the static characteristic of engine effectively, the paper first transforms the nonlinear equations describing engine system characteristics into optimization problem. Then, the improved PSO proposed in the paper is applied to solve the nonlinear and multidimensional problem. The result shows that the improved PSO can analyze effectively.